Abstract:Brain metastases are the most common form of intracranial tumor in adults, however the prognosis remains dismal due to lack of effective treatments and limited understanding of the underlying physiology allowing them to thrive in the foreign cell type environment of the brain. Here we characterized the transcriptomic and epigenomic landscape of 35 human brain metastases obtained from surgical resection using combined scRNA-seq and ATAC-seq spanning the most prevalent primary tumor types including lung adenocarcinoma (n=16), melanoma (n=14), breast (n=10), and renal cell carcinoma (n=6). Varying molecular subtypes across primary tumors were included allowing for a comprehensive analysis between and within primary tumor types. Approximately 200,000 cells total were profiled to a sequencing depth of 50,000 mean reads per cell with 2,500 median genes per cell. ATAC-seq libraries were sequenced to a depth of 25,000 reads per cell with 10-15,000 median fragments per cell. Unbiased clustering and integrated analysis identified various cell types within the tumor, immune microenvironment, and normal brain tissue. Tumor cell identities were confirmed throughout multiple clusters via inferred copy number alteration analyses. Integration with primary tumor atlases highlight differences between cell states of primary and metastatic tumors, revealing shifts of enhancer function in transcriptional regulation of metastatic cells. Overall, our findings describe the genetic and epigenetic states of metastatic cells contributing to the understanding of CNS metastatic colonization.

Epco-42. Single-Cell Multiomic Analysis Of Brain Metastases Across Multiple Primary Tumor Types